A wireless controlled robotic insect with ultrafast untethered running speeds

Insect-scale legged microrobots (less than 5 cm) face a significant challenge: maintaining high running speeds after carrying payloads such as onboard power and control units. While various insect-scale robots demonstrate satisfactory tethered locomotion, untethered performance is severely limited by the weight of these components. Existing approaches utilize walking or running gaits. Walking gaits, exemplified by the Harvard HAMR, require complex actuation and experience significant speed reduction untethered. Running gaits, using body deformation for bouncing, also suffer from substantial speed degradation with payloads. This research addresses this bottleneck by presenting a novel 2-cm legged microrobot that achieves high untethered running speeds comparable to insects, using a unique actuation mechanism and gait.

The authors review existing insect-scale legged microrobots, highlighting the limitations of both walking and running gaits in achieving high-speed untethered locomotion. Walking gaits, such as those used in the Harvard HAMR robot, often require complex actuation mechanisms and suffer significant speed reductions when operating untethered due to the added weight of onboard power and control systems. Running gaits, which utilize body deformation to generate bouncing movements, have also shown substantial speed degradation when payloads are integrated. The authors cite several examples of insect-inspired robots and their limitations in untethered operation, motivating the need for a new approach to achieve high-speed untethered locomotion.

The BHMbot, a 2-cm long, 1760 mg microrobot, is designed with two electromagnetic actuators, two transmissions integrating two front legs, two rear legs, and support frames. The electromagnetic actuators, with a power density exceeding 200 W kg⁻¹, generate vibratory motions. A four-bar linkage transmission converts this reciprocating motion into a leg swing. The front legs are longer than the rear legs, creating a body tilt angle crucial for generating obliquely upward bouncing momentum. The gait is characterized by a series of continuous bouncing cycles. The authors conducted experiments with prototypes of varying body lengths to optimize design parameters such as body length, cantilever width, relative distance between coil and magnet, body tilt angle, and rear leg length. A simplified dynamical model was developed to aid in the optimization process. This involved analyzing the relationship between running speed, payload mass, and various structural parameters. The model uses four generalized coordinates to describe the robot's motion. The optimization aimed at maximizing running speed and load-carrying capacity. Tethered locomotion tests were conducted to evaluate the effect of body length and payload mass on running speed. The authors also fabricated miniaturized power and control electronics (600 mg) using a Bluetooth microcontroller unit, H-Bridge drivers, and a Schmitt-Trigger inverter. This allowed for wireless control of the BHMbot's locomotion trajectories. Untethered locomotion performance was evaluated on various surfaces, measuring linear running speed and turning centripetal acceleration. Cost of Transport (COT) was also calculated to assess energy efficiency. Finally, application scenarios such as sound signal detection, internal inspection of turbofan engines, and drone-assisted transportation were demonstrated.

The BHMbot achieved a remarkable untethered running speed of 17.5 BL s⁻¹ and a turning centripetal acceleration of 65.4 BL s⁻², exceeding the performance of previously reported untethered insect-scale microrobots. The study revealed a unique relationship between running speed and payload mass, where adding a certain optimal mass increased speed before decreasing it. This counterintuitive finding is explained through kinematic and dynamic analysis. Kinematic analysis showed that the increase in bouncing frequency compensated for the decrease in bouncing length. Dynamic analysis revealed that increasing payload mass initially increased the effective power used for locomotion, and that above a certain payload mass, the dissipated power increased, reducing speed. A simplified dynamic model helped to optimize design parameters for maximum running speed and load-carrying capacity. Wireless control was successfully demonstrated, enabling the BHMbot to navigate complex trajectories, including circles, rectangles, letters, and irregular paths across obstacles. The robot's performance was tested on various surfaces and the cost of transport was calculated, showing its energy efficiency. The BHMbot successfully demonstrated several application scenarios, including sound detection, internal engine inspection, and collaboration with a drone for extended range.

The BHMbot's exceptional untethered performance addresses the long-standing challenge of payload-induced speed degradation in insect-scale robots. The findings highlight the importance of considering the interplay between bouncing length and frequency in gait design. The development of a simplified dynamic model for parameter optimization is a significant contribution to microrobotics design. The successful wireless control and application demonstrations showcase the potential of the BHMbot for various real-world applications. The achieved running speed and turning agility of the BHMbot are comparable to those of some insects, demonstrating the feasibility of creating biologically-inspired robots with high mobility. The research further suggests that the proposed actuation mechanism could be adopted by other insect-scale microrobots.

This research presents the BHMbot, a 2-cm legged microrobot achieving ultrafast untethered running speeds and remarkable turning agility through a novel actuation mechanism and gait design. The study's key contributions include the optimization of structural parameters using a simplified dynamic model, achieving wireless trajectory control, and demonstrating successful application in various scenarios. Future research could focus on further miniaturization, exploring alternative actuation mechanisms, enhancing power efficiency, and expanding application capabilities.

The current BHMbot design has limitations in terms of battery life (3 minutes) and its inability to climb inclines beyond a certain angle. The simplified dynamic model used for optimization may not capture all aspects of the complex dynamics of the robot's locomotion. The range of the wireless control is limited by the Bluetooth module's capabilities. Further research is needed to address these limitations and improve the robustness of the robot in more challenging environments.